In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
Make sure that you've downloaded the required human and dog datasets:
Note: if you are using the Udacity workspace, you DO NOT need to re-download these - they can be found in the /data folder as noted in the cell below.
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dog_images.
Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
import numpy as np
from glob import glob
import os
# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
Question 1: Use the code cell below to test the performance of the face_detector function.
human_files have a detected human face? dog_files have a detected human face? Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: 98 percent of the human files had a human face detected. 17 percent of the dog files had a human face detected.
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
faces_detected_human_files = 0
faces_detected_dog_files = 0
for human in human_files_short:
if (face_detector(human)):
faces_detected_human_files += 1
for dog in dog_files_short:
if(face_detector(dog)):
faces_detected_dog_files += 1
print('%d images of humans were detected.' % faces_detected_human_files)
print('%d images of humans were detected in the dog files! :)' % faces_detected_dog_files)
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
In this section, we use a pre-trained model to detect dogs in images.
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
VGG16.eval()
img = Image.open(img_path)
transform = transforms.Compose([transforms.Resize(224), transforms.CenterCrop(224),
transforms.ToTensor(), transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5))])
tensor = transform(img)
tensor = tensor.unsqueeze(0)
if use_cuda:
tensor = tensor.cuda()
classification = VGG16(tensor)
classification_index = torch.argmax(classification, 1).item()
return classification_index # predicted class index
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## Get the index of the picture classification
index = VGG16_predict(img_path)
#If it is in the dog range, return True
if (index >= 151 and index <= 269):
return True
else:
return False
Question 2: Use the code cell below to test the performance of your dog_detector function.
human_files_short have a detected dog? dog_files_short have a detected dog? Answer: No dogs were detected in the human files, and all of the dog files had dogs in them.
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
faces_detected_human_files = 0
faces_detected_dog_files = 0
for human in human_files_short:
if (dog_detector(human)):
faces_detected_human_files += 1
for dog in dog_files_short:
if(dog_detector(dog)):
faces_detected_dog_files += 1
print('%d images of dogs were detected in the human files.' % faces_detected_human_files)
print('%d images of dogs were detected in the dog files.' % faces_detected_dog_files)
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
from PIL import Image
import torchvision.transforms as transforms
InceptionV3 = models.inception_v3(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
InceptionV3 = InceptionV3.cuda()
InceptionV3.eval()
def InceptionV3_predict(img_path):
img = Image.open(img_path)
transform = transforms.Compose([transforms.Resize(299), transforms.CenterCrop(299), transforms.ToTensor(),
transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5))])
tensor = transform(img)
tensor = tensor.unsqueeze(0)
if use_cuda:
tensor = tensor.cuda()
classification = InceptionV3(tensor)
classification_index = torch.argmax(classification, 1).item()
return classification_index # predicted class index
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector_inception(img_path):
## Get the index of the picture classification
index = InceptionV3_predict(img_path)
#If it is in the dog range, return True
if (index >= 151 and index <= 268):
return True
else:
return False
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
faces_detected_human_files = 0
faces_detected_dog_files = 0
for human in human_files_short:
if (dog_detector_inception(human)):
faces_detected_human_files += 1
for dog in dog_files_short:
if(dog_detector_inception(dog)):
faces_detected_dog_files += 1
print('%d images of dogs were detected in the human files.' % faces_detected_human_files)
print('%d images of dogs were detected in the dog files.' % faces_detected_dog_files)
###All of the dogs were detected, but one dog was detected in the human files. They both seem to be very accurate.
###I will use the more accurate VGG16 model for my transfer learning project.
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
![]() |
![]() |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
![]() |
![]() |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
![]() |
![]() |
![]() |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import os
import numpy as np
import torch
from torchvision import datasets
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
%matplotlib inline
from PIL import Image
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
num_workers = 0
batch_size = 20
#Transform for the test and validation data
transform = transforms.Compose([transforms.Resize(224), transforms.CenterCrop(224),
transforms.ToTensor(), transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5))])
#Augmented transform for the test data set
transform_train = transforms.Compose([transforms.Resize(224), transforms.CenterCrop(224),
transforms.RandomRotation(25), transforms.RandomHorizontalFlip(),
transforms.RandomResizedCrop(224),
transforms.ToTensor(), transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5))])
train_data = datasets.ImageFolder('/data/dog_images/train', transform=transform_train)
test_data = datasets.ImageFolder('/data/dog_images/test', transform=transform)
validation_data = datasets.ImageFolder('/data/dog_images/valid', transform=transform)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size,
num_workers=num_workers, shuffle=True)
validation_loader = torch.utils.data.DataLoader(validation_data, batch_size=batch_size,
num_workers=num_workers)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size,
num_workers=num_workers)
loaders_scratch = {'train': train_loader, 'test': test_loader, 'valid': validation_loader}
num_breeds = len(test_data.classes)
print (num_breeds)
#Visualize samples of the data in each set
def visualize(loader, data):
new_iter = iter(loader)
images, labels = next (new_iter)
images = images.numpy()
breeds = data.classes
fig = plt.figure(figsize=(25, 4))
for index in np.arange(20):
ax = fig.add_subplot(2, 20/2, index+1, xticks=[], yticks=[])
image = images[index]
image = image / 2 + 0.5
plt.imshow(np.transpose(image, (1, 2, 0)))
ax.set_title(breeds[labels[index]])
visualize(train_loader, train_data)
visualize(test_loader, test_data)
visualize(validation_loader, validation_data)
Question 3: Describe your chosen procedure for preprocessing the data.
Answer: My code resizes the images to 224 with a resize and center crop. I picked 224 as the size because that is the size that is same size my transfer learning project will use with the VGG16 model. I will be able to compare the difference with the same sized images. I will augment my training images with flips, rotations, and random resized crops to help my network generalize and not overfit the data.
Create a CNN to classify dog breed. Use the template in the code cell below.
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
### TODO: choose an architecture, and complete the class
def __init__(self):
super(Net, self).__init__()
## Convolutional layers
#Image starts at 224 x 224
self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
#Image is 112 x 112
self.conv2 = nn.Conv2d(16, 32, 3, padding=1)
#Image is 56 x 56
self.conv3 = nn.Conv2d(32, 64, 3, padding=1)
#Image is 28 x 28
self.conv4 = nn.Conv2d(64, 128, 3, padding=1)
#Image is 14 x 14
self.conv5 = nn.Conv2d(128, 256, 3, padding=1)
##Pooling layer
self.pool = nn.MaxPool2d(2,2)
##Batch normalization
self.bn1 = nn.BatchNorm2d(16)
self.bn2 = nn.BatchNorm2d(32)
self.bn3 = nn.BatchNorm2d(64)
self.bn4 = nn.BatchNorm2d(128)
self.bn5 = nn.BatchNorm2d(256)
##Linear layers
self.fc1 = nn.Linear(256 * 7 * 7, 2000)
self.fc2 = nn.Linear(2000, 500)
self.fc3 = nn.Linear(500, 133)
def forward(self, x):
## Define forward behavior
x = self.pool(F.relu(self.bn1(self.conv1(x))))
x = self.pool(F.relu(self.bn2(self.conv2(x))))
x = self.pool(F.relu(self.bn3(self.conv3(x))))
x = self.pool(F.relu(self.bn4(self.conv4(x))))
x = self.pool(F.relu(self.bn5(self.conv5(x))))
x = x.view(-1, 256 * 7 * 7)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
# instantiate the CNN
model_scratch = Net()
print (model_scratch)
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer: I am using five convolutional layers to try to recognize the complex patterns of the dog breeds. After each convolutional layer, I am using a batch normalization layer to help optimize the model and improve its performance. Then, I am applying a relu function and a 2 x 2 pooling layer. After the convolutional layers, I flatten the tensor, then it goes through three linear layers. I choose 2000 for the output to the first layer, and 500 as the output of the second layer, because these numbers were between the output of my convolutional layers (12544) and my ultimate output (133) for the number of breeds. After reading questions from others about improving the accuracy of their networks, I decided not to use dropout layers, and found that it improved my network's accuracy. I also used the Adam optimizer and found that it helped with accuracy.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters(), lr=0.002)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# Initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# Initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
# Move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## Find the loss and update the model parameters accordingly
## Record the average training loss
optimizer.zero_grad()
output = model(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
######################
# validate the model #
######################
model.eval()
for batch_idx, (data, target) in enumerate(loaders['valid']):
# Move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
output = model(data)
loss = criterion(output, target)
valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
#train_loss = train_loss/(len(train_loader.dataset))
#valid_loss = valid_loss/(len(valid_loader.dataset))
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch,
train_loss,
valid_loss
))
## TODO: save the model if validation loss has decreased
if (valid_loss < valid_loss_min):
print('Validation loss decreased. Saving model.')
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# return trained model
return model
# train the model
model_scratch = train(30, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# Load trained model again (doing this over multiple settings)
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
# call test function
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
## TODO: Specify data loaders
loaders_transfer = {'train': train_loader, 'test': test_loader, 'valid': validation_loader}
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
#Load model and print output
import torchvision.models as models
import torch.nn as nn
model_transfer = models.vgg16(pretrained=True)
print (model_transfer)
print (use_cuda)
#Freeze training for features layers
for param in model_transfer.features.parameters():
param.requires_grad = False
#Change the final layer in the classifier to check dog breeds.
transfer_classifier = nn.Sequential(nn.Linear(4096, 2000),
nn.ReLU(),
nn.Dropout(.2),
nn.Linear(2000, 1000),
nn.ReLU(),
nn.Dropout(.2),
nn.Linear(1000, 133))
model_transfer.classifier[6] = transfer_classifier
if use_cuda:
model_transfer = model_transfer.cuda()
print (model_transfer.classifier[6])
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer: I chose to use the VGG16 model for this project. I have kept the features layer (where the convolutional layers are) from training by setting its requires_grad to False, so only the layers in the classifier will train. I changed the final layer in the classifier. I changed the linear layer that accepts 4096 inputs to have 2000 outputs. I am applying a relu function to this layer. I have a secong layer that accepts the 2000 outputs from this layer and puts out 1000 ouputs. Finally, I am adding a final layer that will take in the 1000 outputs from the previous layer as inputs and will produce 133 outputs for the 133 breeds that the network will need to classify. At first, a tried this with no dropout and only two layers, but I only got 48 percent accuracy. Adding an additional layer to help give the netork room to learn and adding dropout to keep it from overfitting increased my accuracy to 70 percent.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
import torch.optim as optim
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.Adam(model_transfer.classifier.parameters(), lr=0.001)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# train the model
model_transfer = train(20, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# Load the model that got the best validation accuracy
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
#Load saved model
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
#Find how files are named
import os
list_dir = os.listdir('/data/dog_images/valid')
from pprint import pprint
pprint (list_dir)
list_dir = os.listdir('/data/dog_images/valid/103.Mastiff')
pprint(list_dir)
import os, random
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = list()
for item in validation_data.classes:
item = item.replace("_", " ")
items = item.split(".")
item = items[1]
class_names.append(item)
print (item)
def predict_breed_transfer(img_path):
model_transfer.eval()
# load the image and return the predicted breed
img = Image.open(img_path)
transform = transforms.Compose([transforms.Resize(224), transforms.CenterCrop(224),
transforms.ToTensor(), transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5))])
tensor = transform(img)
tensor = tensor.unsqueeze(0)
if use_cuda:
tensor = tensor.cuda()
output = model_transfer(tensor)
classification_index = torch.argmax(output, 1).item()
return classification_index
def get_breed_image_path(index):
#Build the path to a random photo of the dog breed and return it.
folder_path = '/data/dog_images/valid/'
breed_path = validation_data.classes[index]
folder_path = folder_path + breed_path
photo_file = random.choice(os.listdir(folder_path))
full_photo_path = folder_path + '/' + photo_file
return full_photo_path
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!

### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
## Handle cases for a human face, dog, and neither
is_dog = dog_detector(img_path)
is_person = face_detector(img_path)
img = Image.open(img_path)
fig = plt.figure(figsize=(10, 10))
axs = fig.subplots(2)
axs[0].set_xticks([])
axs[0].set_yticks([])
axs[1].set_xticks([])
axs[1].set_yticks([])
if (is_dog):
fig.suptitle('This is a dog.')
axs[0].imshow(img)
axs[0].set_title('It is a(n) ...')
elif (is_person):
fig.suptitle('This is a person.')
axs[0].imshow(img)
axs[0].set_title('But if it was a dog, it would be a(n)...')
else:
fig.suptitle('Error. This is not a person or a dog. ')
axs[0].imshow(img)
axs[0].set_title('But if it was a dog, it would be a(n)...')
predicted_breed_index = predict_breed_transfer(img_path)
breed_image_path = get_breed_image_path(predicted_breed_index)
breed_image = Image.open(breed_image_path)
axs[1].imshow(breed_image)
axs[1].set_title(class_names[predicted_breed_index])
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer: The algorithm output is better than I expected. You can really see the visual similarities between the photos given and the breeds predicted, whether the photo is of a person, a dog, or another animal. It mistook the photo of the Husky I gave it, although the breed that it predicted looks very similar. It guessed the breeds of the collie and Yorkie photos that I gave it. The dog detector works well and was not tricked by a fox. I like the pictures that it chose for the people and animals that I tested it on (particularly the one it chose for Mr. Spock!)
Points for improvement:
I could continue tweeking the transfer_classifier to try to get higher accuracy.
I could format the breed names to be consistent by making sure that in breeds that are multiple words, all the words are capitalized.
I could check the first letter of the breed name and change the message to "But if it was a dog, it would be a(n)" to ending with "a" or "an" based on whether the breed name starts with a vowel.
I could maybe change the text size of "This is a person" to be larger.
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
run_app('images/galadriel.jpg')
run_app('images/husky.jpg')
run_app('images/flying_fox.JPG')
run_app('images/spock.jpg')
run_app('images/yorkie.jpg')
run_app('images/hippo.jpg')
run_app('images/bilbo.jpg')
run_app('images/thorin.jpg')
run_app('images/collie.jpg')
run_app('images/fox.jpg')